Service Level Management
Primary Transaction Form: SLM:Measurement Primary Escalation: SLM:EventSchedule:TAD_PollingEscalation Configure SLM agreements in the Service Level Management Console, which may be launched from the Home Page. Installing You do not need to install Service Level Management Collector unless you will be configuring SLM to measure non-ITSM records. Asset Availability SLM may be used to measure Asset Availability. This will create an SLM:Measurement record whenever an outage record (AST:CI Unavailability) is created for an asset, and it will update the measurement record when the asset outage is resolved. Reading the Measurement Record The measurement record will keep the outage start time in the "OverallStartTime" field on the "Misc" tab, and the outage end time on the "UpStartTime" field on the same tab. Configuring Asset Availability Service Targets The availability is driven from the Outage record. This example uses the 'Status-ACI' field as the key indicator of the outage. There are some nice templates to help you configure an Asset Availability record. *Goal Type: Asset Availability *Measurement Template: Asset Availability Reported CI Status **Available When: 'Status-ACI' = "Deployed" **Unavailable When: 'Status-ACI' = "Down" *Lifecycle Interval: 1 Month Understanding the dashboard On the Service Level Management Console, there is a "Dashboard" tab. Under that tab, there are many sub-tabs. Select the "Service Targets" sub-tab. The pie chart on the right shows the availability of the measured asset outages at this moment (not yet sure why this is relevant). An asset is down when the status is set to "Down" and an Asset Outage record is created for that asset in the "Current Unavailability" status. The table in the lower-left corner shows the outages. If any outages are currently in effect (have a status of Current Unavailability"), they will be red. Highlighting a record in the left table will refresh the right table, where you can see a running tally of how long the asset was unavailable (click the "Refresh" button to recalculate). Ending the outage To end the outage, open the asset, go to the Outage tab, select the outage and click "View". Set the "Reported CI Status" to "Deployed". Set the Assignment Status to Completed. Type $TIMESTAMP$ into the Actual End Date record and hit "Save". Move the "Asset Status" back to "Deployed". Incidents SLM is most commonly used to measure Incident Reponse and Resolution time. It is used to send warning messages to the assignee, assignee group, or specific email addresses when the times pass a certain threshold, and they can be used to measure compliance. To configure for Incident Management, go to the left nav and select Folders-->Standard Services-->Incident Management Use the "Service Targets" tab. Work hours may be configured using Business Calc Time to allow for non 24x7 SLAs Milestone Actions Milestone actions may be used to email individuals when the SLA Due Date nears. This can be done in one of two ways. First, you can build a "Notify or Email" action. These are very reliable but do not use the Notification Engine, and thus users may not unsubscribe from the messages. Secondly, you may use a "Set Value" action. Simply set the z1D Action field to one of the values below.. Accepted values for z1D Action field *ESCALATEDBYOLA *ESCALATEDBYSLA *ESCALATEMGRASSIGN *ESCALATEOWNER *ESCALATEMGROWNER Filters map the z1D Action to Notification Templates The z1D Action values map to specific notification templates in the SYS:Notification Messages form by specific filters. The z1D Action field is defined in the filter Run-If Qualification, and the Notification Message Tag is set to the z1D Notification Message Tag in the Set Fields action. Also, note that the filter "HPD:INC:BypassToNotify_MilestoneEscalate" sets the escalated flag, increments the Total Escalation Level, and skips to execution order 865. Any customization adding new z1D Action fields should also modify the qualification of this filter. The filters then call the HPD:INC:NotificationGenerator guide, which in turn calls the single filter "HPD:INC:NotificationGenerator_899_PNPC`!", which creates a record in NTE:SYS-NT Process Control BMC Defect asking for explanations of the z1D Action Values: SW00268723 http://www.mail-archive.com/arslist@arslist.org/msg37875.html Adding Work Notes # Impliment the customization: Create Work Note from Incident # Add a "Set Value" milestone action to set the z1D Action field to "Add Work Log" and the z1D_Note field to the text of the work note. Milestone Notification Recipients *$Assigned Group$ *$Assignee Login ID$ (not $Assignee$) Allow Service Target to Reopen By default, an SLA that is met will start processing again if the ticket is modified out of its close criteria. For example, if an SLA is met when an Incident reaches the status of "Resolved", and the incident is later modified back to "In Progress", any open SLAs will begin to process again. If this is not the desired behavior, it can be turned off by setting the "Allow Service Target to Reopen" field to "No". This is defined on the Service Target. Service Target Groups By putting the service targets into service target groups, then one service target may "take over" for another when the conditions of the ticket change. For example, if an incident changes priority, and the SLA Service Targets are defined by priority, then the new service target will start at the same time as the old service target. Create a service target group using the SLM:ConfigSLAGroup form. Application Administration Console --> Service Level Management --> Configure Application Settings --> Service Target Groups You can also add service targets to a group using the "Measurement" tab of the service target. Templates You can configure a group of similar SLAs faster by using the templates. To create a template to email the assignee of an incident: #From the Application Administration Console, go to Custom Configuration tab --> Service Level Management --> Configure Application Settings --> Templates #The configure Templates form appears. In "Show Templates For", select "Action" and click the "Create" button #Set the following in the "Create Action Template" dialog that appears: ##Used By: Request Based ##Applies To: Incident ##Action Type: Alert or Email #Click "OK" #Complete the template. Forms that store the templates *Actions: SLM:RuleAction Troubleshooting Service Targets never finish building - say "Build in progress" Service targets usually build in seconds, but sometimes they take longer, especially if the server is slow. Try waiting it out, but if hours pass, here are some things that may help. *Check if slmbrsvc process is running. Windows (slmbrsvc.exe) *Check if there is a line in armonitor.cfg for starting up slmbrsvc. *Make sure the following are set to "F" in the ar.cfg file: **Approval-Server-Suspended: F **Assignment-Engine-Suspended: F **Business-Rules-Engine-Suspended: F **CMDB-Service-Suspended: F **Reconciliation-Engine-Suspended: F **SLM-Collector-Suspended: F Customizations *SLM-Response Notifications via Notification Engine Category:ITSM Category:Service Level Management